Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
关于 JoVE
概览领导团队博客JoVE 帮助中心
作者
出版流程编辑委员会范围与政策同行评审常见问题投稿
图书馆员
用户评价订阅访问资源图书馆顾问委员会常见问题
研究
JoVE JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

相关概念视频

Doppler Effect - II01:05

Doppler Effect - II

3.3K
The Doppler effect has several practical, real-world applications. For instance, meteorologists use Doppler radars to interpret weather events based on the Doppler effect. Typically, a transmitter emits radio waves at a specific frequency toward the sky from a weather station. The radio waves bounce off the clouds and precipitation and travel back to the weather station. The radio frequency of the waves reflected back to the station appears to decrease if the clouds or precipitation are moving...
3.3K

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序
Same author

Beyond the Cuff: State-of-the-Art on Cuffless Blood Pressure Monitoring.

Sensors (Basel, Switzerland)·2026
Same author

Development of open radio access networks (O-RAN) for real-time robotic teleoperation.

Communications engineering·2025
Same author

Memetic Salp Swarm Algorithm for economic load dispatch problems.

Scientific reports·2025
Same author

Multimodal Sensing-Enabled Large Language Models for Automated Emotional Regulation: A Review of Current Technologies, Opportunities, and Challenges.

Sensors (Basel, Switzerland)·2025
Same author

Slotted ALOHA Based Practical Byzantine Fault Tolerance (PBFT) Blockchain Networks: Performance Analysis and Optimization.

Sensors (Basel, Switzerland)·2024
Same author

Artificial intelligence enabled smart mask for speech recognition for future hearing devices.

Scientific reports·2024

相关实验视频

Updated: May 28, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

2.6K

雷达信号处理及其对深度学习驱动的人类活动识别的影响

Fahad Ayaz1, Basim Alhumaily1, Sajjad Hussain1

  • 1James Watt School of Engineering, University of Glasgow, Glasgow G12 8QQ, UK.

Sensors (Basel, Switzerland)
|February 13, 2025
PubMed
概括

雷达技术与卷积神经网络 (CNN) 结合,提高了人类活动识别 (HAR). 带有STFT预处理的MobileNetV2实现了96.30%的准确性,平衡实时应用程序的效率和性能.

关键词:
计算成本是计算成本.深度学习是一种深度学习.人类活动分类人类活动分类雷达领域表示 雷达领域表示转移学习转移学习

更多相关视频

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
06:49

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Published on: December 11, 2015

8.9K
A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis
05:41

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

9.3K

相关实验视频

Last Updated: May 28, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

2.6K
Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
06:49

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Published on: December 11, 2015

8.9K
A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis
05:41

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

9.3K

科学领域:

  • 雷达信号处理 雷达信号处理
  • 机器学习 机器学习
  • 计算机视觉 计算机视觉

背景情况:

  • 人类活动识别 (HAR) 对智能安全和医疗保健至关重要.
  • 雷达技术为HAR提供了一种保护隐私的替代方案.
  • 将先进的信号处理与深度学习相结合是改善HAR的关键.

研究的目的:

  • 调查CNN与雷达信号处理的整合,以提高HAR.
  • 评估不同的雷达地图生成技术和CNN架构.
  • 为了确定实时,资源受限的HAR应用程序的最佳配置.

主要方法:

  • 使用了三种2D雷达处理技术:范围-FFT时间范围地图,STFT时间多普勒地图和SPWVD地图.
  • 评估了四个CNN架构:VGG-16,VGG-19,ResNet-50和MobileNetV2. 这四个架构都在进行评估.
  • 分析了12个CNN和预处理配置,以获得准确性和计算效率.

主要成果:

  • 带有STFT预处理的MobileNetV2实现了96.30%的准确性.
  • 这种配置表现出高计算效率,推断时间为2.57ms.
  • 谱图生成时间为220毫秒,适合实时处理.

结论:

  • 雷达生成的地图作为HAR的有效视觉数据,确保隐私.
  • 该研究强调了预处理复杂性和识别精度之间的权衡.
  • 最佳的配置,如STFT的MobileNetV2,使得基于雷达的HAR在边缘计算和AR中得到更广泛的应用.